Institution
University of East Anglia
Education•Norwich, Norfolk, United Kingdom•
About: University of East Anglia is a education organization based out in Norwich, Norfolk, United Kingdom. It is known for research contribution in the topics: Population & Climate change. The organization has 13250 authors who have published 37504 publications receiving 1669060 citations. The organization is also known as: UEA.
Topics: Population, Climate change, Randomized controlled trial, Health care, Psychological intervention
Papers published on a yearly basis
Papers
More filters
••
TL;DR: Through extensive experimentation on 72 datasets, it is demonstrated that the simple collective formed by including all classifiers in one ensemble is significantly more accurate than any of its components and any other previously published TSC algorithm.
Abstract: Recently, two ideas have been explored that lead to more accurate algorithms for time-series classification (TSC) First, it has been shown that the simplest way to gain improvement on TSC problems is to transform into an alternative data space where discriminatory features are more easily detected Second, it was demonstrated that with a single data representation, improved accuracy can be achieved through simple ensemble schemes We combine these two principles to test the hypothesis that forming a collective of ensembles of classifiers on different data transformations improves the accuracy of time-series classification The collective contains classifiers constructed in the time, frequency, change, and shapelet transformation domains For the time domain, we use a set of elastic distance measures For the other domains, we use a range of standard classifiers Through extensive experimentation on 72 datasets, including all of the 46 UCR datasets, we demonstrate that the simple collective formed by including all classifiers in one ensemble is significantly more accurate than any of its components and any other previously published TSC algorithm We investigate alternative hierarchical collective structures and demonstrate the utility of the approach on a new problem involving classifying Caenorhabditis elegans mutant types
330 citations
••
TL;DR: The SML paradigm is discussed, taking into account physicochemical and biological characteristics that define SML structure and function, and previously unpublished time series data on bacterioneuston composition and SML surfactant activity immediately following physical SML disruption are presented.
330 citations
••
TL;DR: It is shown that for a migratory bird this process can apply on a country-wide scale with consequences for both survival and timing of arrival on the breeding grounds (an indicator of reproductive success); the buffer effect can be a major process influencing large-scale population regulation of migratory species.
Abstract: Buffer effects occur when sites vary in quality and fluctuations in population size are mirrored by large changes in animal numbers in poor-quality sites but only small changes in good-quality sites. Hence, the poor sites ‘buffer’ the good sites1,2, a mechanism that can potentially drive population regulation if there are demographic costs of inhabiting poor sites. Here we show that for a migratory bird this process can apply on a country-wide scale with consequences for both survival and timing of arrival on the breeding grounds (an indicator of reproductive success3,4). The Icelandic population of the black-tailed godwit, Limosa limosa islandica, wintering in Britain has increased fourfold since the 1970s (ref. 5) but rates of change within individual estuaries have varied from zero to sixfold increases. In accordance with the buffer effect, rates of increase are greater on estuaries with low initial numbers, and godwits on these sites have lower prey-intake rates, lower survival rates and arrive later in Iceland than godwits on sites with stable populations. The buffer effect can therefore be a major process influencing large-scale population regulation of migratory species.
330 citations
••
TL;DR: In this paper, the authors describe an experiment in which the position of scientists with respect to flood risk management is fundamentally changed, where the event, flooding, is given agency in the experiment, and reveal a deep and distributed understanding of flood hydrology across all experts, certified and uncertified.
Abstract: In this paper, we describe an experiment in which the position of scientists with respect to flood risk management is fundamentally changed. Building on a review of three very different approaches to engaging the public in science, we contrast the normal way in which science is used in flood risk management in England and Wales with an experiment in which knowledge regarding flooding was co-produced. This illustrates a way of working with experts, both certified (academic natural and social scientists) and non-certified (local people affected by flooding), for whom flooding is a matter of concern, and where the event, flooding, is given agency in the experiment. We reveal a deep and distributed understanding of flood hydrology across all experts, certified and uncertified, involved in the experiment. This did not map onto the conventional dichotomy between ‘universal’ scientific expertise and ‘local’ lay expertise. By working with the event we harnessed, produced and negotiated a new and collective sense of knowledge, sufficient in our experiment to make a public intervention in flood risk management in our case-study location. The manner in which the academic scientists involved in the practice of their science were repositioned was radical as compared with normal scientific method. It was also radical for a more fundamental reason: the purpose of our experiment became as much about creating a new public capable of making a political intervention in a situation of impasse, as it was about producing the solution itself. The practice of knowledge generation, the science undertaken, worked with the hybridisation of science and politics rather than trying to extract science from it.
330 citations
••
12 Aug 2012TL;DR: This work describes a means of extracting the k best shapelets from a data set in a single pass, and then uses these shapelets to transform data by calculating the distances from a series to each shapelet.
Abstract: The problem of time series classification (TSC), where we consider any real-valued ordered data a time series, presents a specific machine learning challenge as the ordering of variables is often crucial in finding the best discriminating features. One of the most promising recent approaches is to find shapelets within a data set. A shapelet is a time series subsequence that is identified as being representative of class membership. The original research in this field embedded the procedure of finding shapelets within a decision tree. We propose disconnecting the process of finding shapelets from the classification algorithm by proposing a shapelet transformation. We describe a means of extracting the k best shapelets from a data set in a single pass, and then use these shapelets to transform data by calculating the distances from a series to each shapelet. We demonstrate that transformation into this new data space can improve classification accuracy, whilst retaining the explanatory power provided by shapelets.
329 citations
Authors
Showing all 13512 results
Name | H-index | Papers | Citations |
---|---|---|---|
George Davey Smith | 224 | 2540 | 248373 |
Nicholas J. Wareham | 212 | 1657 | 204896 |
Cyrus Cooper | 204 | 1869 | 206782 |
Kay-Tee Khaw | 174 | 1389 | 138782 |
Phillip A. Sharp | 172 | 614 | 117126 |
Rory Collins | 162 | 489 | 193407 |
William J. Sutherland | 148 | 966 | 94423 |
Shah Ebrahim | 146 | 733 | 96807 |
Kenneth M. Yamada | 139 | 446 | 72136 |
Martin McKee | 138 | 1732 | 125972 |
David Price | 138 | 1687 | 93535 |
Sheila Bingham | 136 | 519 | 67332 |
Philip Jones | 135 | 644 | 90838 |
Peter M. Rothwell | 134 | 779 | 67382 |
Ivan Reid | 131 | 1318 | 85123 |